The present invention relates to a voltage detector for detecting the voltage developing in a selected area of an object to be measured such as an electric circuit. In particular, the present invention relates to a voltage detector of the type that detects voltage by making use of the change in light polarization that occurs in accordance with the voltage developing in a selected area of an object to be measured.
Various voltage detectors have been used to detect the voltage developing in a selected area of objects to be measured such as electric circuits. Conventional voltage detectors are roughly divided into two types: in one type, the probe is brought into contact with a selected area of an object to be measured and the voltage developing in that area is detected; and in the other type, the probe does not make contact with a selected area of an object to be measured and instead an electron beam is launched into that area and the voltage developing in it is detected.
Voltage changes rapidly in fine-line portions of objects such as integrated circuits that are small and complicated in structure, and a strong need exists in the art for detecting such rapidly changing voltage with high precision without affecting the fine-line portions. However, this need has not been fully met by the prior art voltage detectors. With devices of the type that detects voltage by bringing the probe into contact with a selected area of an object to be measured, it is difficult to attain direct contact between the probe and a fine-line portion of the object of interest such as an integrated circuit. Even if this is successfully done, it has been difficult to correctly analyze the operation of the integrated circuit solely on the basis of the voltage information picked up by the probe. A further problem involved is that contact by the probe can cause a change in the operation of the integrated circuit. Voltage detectors of the type that employs an electron beam has the advantage that they are capable of voltage detection without bringing the probe into contact with an object to be measured. However, the area to be measured with such voltage detectors has to be placed in vacuum and its surface must be exposed at that. In addition, the area to be measured is prone to be damaged by electron beams.
The prior art voltage detectors have a common problem in that they are unable to operate quickly enough to follow rapid changes in voltage and hence fail to achieve precise detection of voltages that change rapidly as in integrated circuits.
With a view to solving these problems, it has been proposed by two of the present inventors (Japanese Patent Application No. 137317/1987 filed on May 30, 1987) that voltage be detected by making use of the polarization of a light beam that changes with the voltage developing in a selected area of an object to be measured.
A voltage detector operating on this principle is schematically shown in FIG. 6. The detector generally indicated by 50 is composed of the following components: an optical probe 52; a CW (Continuous-Wave) light source 53 typically in the form of a laser diode; an optical fiber 51 for guiding a light beam from the CW light source 53 into an optical probe 52 via a condenser lens 60; an optical fiber 92 for guiding reference light from the optical probe 52 into a photoelectric converter 55 via a collimator 90; an optical fiber 93 for guiding output light from the optical probe 52 into a photoelectric converter 58 via a collimator 91; and a comparator circuit 61 for comparing the electric signals form the photoelectric converters 55 and 58.
The optical probe 52 is filled with an electro-optic material 62 such as an optically uniaxial crystal of lithium tantalate (LiTaO.sub.3). The tip 63 of the electro-optic material 62 is worked into a frustoconical shape. The optical probe 52 is surrounded with a conductive electrode 64 and has at its tip 63 a coating of reflecting mirror 65 in the form of a thin metal film or a multilayered dielectric film.
The optical probe 52 further includes the following components: a collimator 94; condenser lenses 95 and 96; a polarizer 54 for selectively extracting a light beam having a predetermined polarized component from the light beam passing through the collimator 94; and a beam splitter 56 that splits the extracted light beam from the polarizer 54 into reference light and input light to be launched into the electro-optic material 62 and which allows the output light emerging from the electro-optic material 62 to be directed into an analyzer 57. The reference light is passed through the condenser lens 95 and thence launched into the optical fiber 92, whereas the output light emerging from the electro-optic material 62 is passed through the condenser lens 96 and thence launched into the optical fiber 93.
Voltage detection with the system shown in FIG. 6 starts with connecting the conductive electrode 64 on the circumference of the optical probe 52 to a predetermined potential, say, the ground potential. Then, the tip 63 of the probe 52 is brought close to the object to be measured such as an integrated circuit (not shown), whereupon a change occurs in the refractive index of the tip 63 of the electro-optic material 62 in the probe 52. Stated more specifically, the difference between refractive indices for an ordinary ray and an extraordinary ray in a plane perpendicular to the light-traveling direction will change in the optically uniaxial crystal.
The light beam issuing from the light source 53 passes through the condenser lens 60 and is guided through the optical fiber 51 to be directed into the collimator 94 in the optical probe 52. The beam is polarized by the polarizer 54 and a predetermined polarized component having intensity I is launched into the electro-optic material 62 in the optical probe 52 via the beam splitter 56. Each of the reference light and the input light, which are produced by passage through the beam splitter 56, has an intensity of I/2. As already mentioned, the refractive index of the tip 63 of the electro-optic material 62 varies with the voltage on the object being measured, so the input light launched into the electro-optic material 62 will experience a change in the state of its polarization at the tip 63 in accordance with the change in the refractive index of the latter. The input light is then reflected from the reflecting mirror 65 and makes a return trip through the electro-optic material 62, from which it emerges and travels back to the beam splitter 56. If the length of the tip 63 of the electro-optic material 62 is written as the state of polarization of input light launched into that material will change in proportion to the difference between refractive indices for an ordinary ray and an extraordinary ray and to the length 2l as well. The output light sent back into the beam splitter 56 is thence directed into the analyzer 57. The intensity of the output light entering the analyzer 57 has been decreased to I/4 as a result of splitting with the beam splitter 56. If the analyzer 57 is designed in such a way as to transmit only a light beam having a polarized component perpendicular to that extracted by the polarizer 54, the intensity of output light that is fed into the analyzer 57 after experiencing a change in the state of its polarization is changed from I/4 to (I/4)sin.sup.2 [(.lambda./2)V/V.sub.o ] in the analyzer 57 before it is further fed into the photoelectric converter 58. In the formula expressing the intensity of output light emerging from the analyzer 57, V is the voltage developing in the object to be measured, and V.sub.O is a half-wave voltage.
In the comparator circuit 61, the intensity of reference light produced from the photoelectric converter 55, or I/2, is compared with the intensity of output light produced from the other photoelectric converter 58, or (I/4)sin.sup.2 [(.lambda./2)V/V.sub.o ].
The intensity of output light, or (I/4)sin.sup.2 [(.pi./2)V/V.sub.O ], will vary with the change in the refractive index of the tip 63 of the electro-optic material 62 that occurs as a result of the change in voltage. Therefore, this intensity can be used as a basis for detecting the voltage developing in a selected area of the object to be measured, say, an integrated circuit.
As described above, in using the voltage detector 50 shown in FIG. 6, the tip 63 of the optical probe 52 is brought close to the object to be measured and the resulting change in the refractive index of the tip 63 of the electro-optic material 62 is used as a basis for detecting the voltage developing in a selected area of the object of interest. Therefore, the voltage developing in fine-line portions of a small and complicated object such as an integrated circuit which are difficult to be contacted by a probe or which cannot be contacted by the same without affecting the voltage being measured can be effectively detected by the detector 50 without bringing the optical probe 52 into contact with such fine-line portions. If desired, a pulse light source such as a laser diode that produces light pulses of a very narrow pulse width may be used as a light source to ensure that rapid changes in the voltage on the object to be measured are sampled at extremely short time intervals. Rapid changes in the voltage on the object of interest can be measured with a very high time resolution by using a CW light source and a quick-response detector such as a streak camera. Either method is capable of precision detection of rapid changes in voltage.
While various materials are used as electro-optic material 62 in the optical probe 52 for the voltage detector 50 shown in FIG. 6, most of them are birefringent under natural conditions. For example, an optically uniaxial crystal of lithium tantalate (LiTaO.sub.3) produces a phase retardation on account of birefringence that occurs spontaneously in proportion to the difference between the refractive index for an ordinary ray n.sub.o and the refractive index for an extraordinary ray n.sub.e. This phase retardation occurs even when no voltage is applied to an optically uniaxial crystal and the state of light polarization will change on account of this phase retardation. Therefore, if one wants to detect the voltage on an object of interest precisely by selectively extracting the change in light polarization that results from a change in refractive index upon voltage application, he has to employ a certain means for canceling the phase retardation caused by spontaneous birefringence.
In order to detect the voltage in a selected area of an object of interest with high sensitivity, it is also necessary to reject a dc component from the light emerging from the analyzer 57. Stated more specifically, if the voltage to be measured that develops in an area of interest is a rapidly changing voltage superposed on a dc voltage, one who wants to selectively detect the rapidly changing voltage component must eliminate the change in the state of light polarization that depends on the dc voltage.
FIG. 7 shows the light intensity I vs. voltage V profile for the case where the analyzer 57 into which the output light emerging from the electro-optic material 62 is introduced is disposed at right angle with respect to the polarizer 54. At operating point Pl on the curve shown in FIG. 7, the change in the intensity of output light, .DELTA.I.sub.1, is linear to the change in voltage V.sub.1. However, because of the presence of a dc component I.sub.DC in the intensity of output light I, the detector is unable to selectively detect the change in the intensity of output light .DELTA.I.sub.1 with high sensitivity. It is therefore necessary to convert the intensity I of light output from the analyzer 57 in such a way that it will not contain any of the dc component I.sub.DC although the change in light intensity .DELTA.I.sub.2 is not linear to the change in voltage .DELTA.V.sub.2 as depicted by operating point P2 on the curve shown in FIG. 7.
To meet this need, a voltage detector having the construction shown in FIG. 8 and which is generally indicated by 70 was previously proposed by the present inventors.
The optical probe 17 for the voltage detector 70 shown in FIG. 8 differs from the probe 52 shown in FIG. 6 in that it has a phase compensator 16 between the beam splitter 56 and the electro-optic material 62. By properly adjusting this phase compensator 16, the change in the state of polarization of output light produced from the electro-optic material 62 can be reduced to such a level that it originates only from the voltage applied to the electro-optic material 62. In other words, in detecting the voltage in a selected area of an object to be measured, the phase compensator 16 may be adjusted in such a way that in the absence of any voltage applied to the electro-optic material 62, the polarized component of the output light produced from said material is directed perpendicular to that of input light (launched into that material), and with the least intensity of output light being produced from the analyzer 57. By adjusting the phase compensator 16 in this way, it becomes possible to cancel the change in the state of polarization of output light that is caused by the phase retardation due to the spontaneous birefringence occurring in the electro-optic material 62.
The phase compensator 16 used for this purpose may be a Soleil compensator capable of mechanical adjustment or a Babinet compensator which is suitable for the case where a small-diameter spot is produced by light beam. A Soleil compensator is equipped with a sliding adjustment mechanism that allows for adjustment from the outside of the optical probe 17. Before detection of the voltage in a selected area of an object to be measured, the adjustment mechanism is operated, usually manually, with no voltage applied to the electro-optic material 62, so as to eliminate the effect of phase retardation caused by spontaneous birefringence.
If the phase compensator 16 is of the voltage adjustable type, the dc component I.sub.DC can be rejected from the intensity of output light from the analyzer 57 by first applying a given voltage to the compensator 16 and then adjusting it manually in such a way that it will operate at point P2 on the curve shown in FIG. 7.
The problem is that the phase retardation caused by spontaneous birefringence occurring in the electro- o optic material 62 is temperature-dependent and varies with the change in the temperature of the optical probe 17, or the environment in which the electro-optic material 62 is used. The intensity of output light at operating point P2 on the curve shown in FIG. 7 also varies with temperature change. It should particularly be mentioned that at operating point P2 the intensity of output light varies nonlinearly, so that a variation in the intensity of output light on account of temperature change can cause a significant effect on the precision of voltage detection. Therefore, even if the phase compensator 16 used in the voltage detector 70 shown in FIG. 8 is manually adjusted to a predetermined value prior to voltage detection, the initial setting will become no longer appropriate in the phase of subsequent changes in temperature and correct results cannot be attained over the entire period of voltage detection.